AI & ML

yeachan-heo Bundle

Create and execute durable repo-native multi-goal plans over GJC goal mode artifacts.

yeachan-heo Bundle

Socratic deep interview with mathematical ambiguity gating before explicit execution approval

yeachan-heo Skill

Adds domain design guidance for ralplan planner phase.

atrislabs Skill

Dispatch work to an installed terminal agent or named Atris engine profile. Supports Atris Fast, Claude, Codex, Cursor, Fable, Composer, Haiku, Devin, Grok, Antigravity (agy), and opencode. Triggers on: use codex, use cursor, use devin, use grok, use agy, use antigravity, use gemini, gemini session, use fable, use claude, use opencode, use atris, engine, dispatch to, worker agent, second opinion build.

atrislabs Skill

Create and manage AI team members using the MEMBER.md format. Use when the user wants to define a new AI role, set up a team member, create an agent persona, or work with team/MEMBER.md files.

atrislabs Skill

Use the local Atris wiki as durable project memory. Use when ingesting source material, answering questions from saved knowledge, finding stale pages, or keeping agent context sharp.

atrislabs Skill

Install and use Atris as an on-disk brain for agent work. Use when starting in any repo, navigating code, claiming tasks, proving work, or running missions that should survive the session.

peterbamuhigire Skill

Use when performing a read-only safety review of a new or changed skill for credential harvesting, unsafe installers, hidden mutation, excessive permissions, unknown tools, or instruction conflicts; use skill-writing for structural quality and source-verification for factual claims.

peterbamuhigire Bundle

Use when an authorised implementation or repository change requires affected project documentation, agent guidance, architecture notes, API/database docs, examples, or status records to be reconciled with the current code; use skill-writing for skill contracts rather than project documentation.

peterbamuhigire Skill

Use when auditing or improving the business-plan engine or any business plan, strategy, model, pitch, nonprofit plan, or execution framework it produces.

bbrysonelite-max Bundle

Use when a declared threat model or risk register needs evidence that each recorded disposition and mitigation is implemented before release.

involvex Bundle

Build applications powered by GitHub Copilot using the Copilot SDK. Use when creating programmatic integrations with Copilot across Node.js/TypeScript, Python, Go, or .NET. Covers session management, custom tools, streaming, hooks, MCP servers, BYOK providers, session persistence, custom agents, skills, and deployment patterns. Requires GitHub Copilot CLI installed and a GitHub Copilot subscription (unless using BYOK).

involvex Bundle

Implements Manus-style file-based planning for complex tasks. Creates task_plan.md, findings.md, and progress.md. Use when starting complex multi-step tasks, research projects, or any task requirin...

jovanipink Bundle

Prepare a precise handoff for continuing work in another session, client, person, or agent using exact revisions, current state, evidence, decisions, blockers, and authority boundaries. Invoke explicitly when continuity matters and the recipient must not rediscover or overstate progress.

jovanipink Bundle

Guide a person through configuration steps that require their account, device, approval, credential, or external interface while verifying each observable result. Invoke explicitly when the agent cannot or should not perform the protected step itself.

jovanipink Bundle

Coordinate authorized parallel agents on independent bounded tasks with explicit ownership, evidence contracts, cost awareness, and final reconciliation. Invoke explicitly when parallel work can reduce latency without creating edit collisions or delegating irreversible decisions.

jovanipink Bundle

Coordinate an explicitly requested source-to-publication learning lifecycle across research, evidence tracing, model or forecast evaluation, corrections, and retrospective findings. Invoke explicitly when the user wants traceable knowledge objects or a public-safe research index; do not use for a one-off search, ordinary summary, or automatic publication.

tencentcloudbase Bundle

CloudBase official HTTP API client guide. This skill should be used when backends, scripts, or non-SDK clients must call CloudBase platform APIs over raw HTTP instead of using a platform SDK or MCP management tool.

chillicream Skill

Send and receive mail between coding agents with `nitro agent mail`: local-first, per-workspace messaging for multi-agent coordination. Use when messaging or replying to another agent, draining an inbox, or waiting for new mail.

chillicream Skill

Create and manage tasks with `nitro agent tasks`, the local-first, dependency-aware issue tracker for coding agents. Use when creating or triaging tasks, wiring dependencies, finding ready work, or updating and closing what you worked on.

chillicream Skill

Save and recall shared, durable agent memory with `nitro agent memory`: knowledge that must outlive the session -- preferences, decisions, domain facts, references -- readable by every agent in the workspace. Use when the user says "remember this", when a session should start from what the workspace already knows, or when knowledge worth keeping surfaces mid-task.

chillicream Bundle

Plan a large, fuzzy feature as a shared map of decision tickets in nitro agent tasks, resolve them one per session until the way is clear, then cut the implementation-ready task graph. Use when the user says "wayfind", "chart a map", "work the map", "next decision", "plan this like a map", invokes /nitro-wayfinder, or brings an effort too big and foggy for one agent session. Not for work that fits one session; create tasks with nitro agent tasks directly for that (see the nitro-task skill).

chillicream Bundle

Authoring guide and conventions for ChilliCream agent skills. Use whenever the user wants to create a new skill, edit an existing one, improve a skill's description, or asks how to structure a SKILL.md file. Triggers on phrases like "write a skill", "new skill", "author a skill", "improve this skill", "skill description", "SKILL.md", or when editing any file under skills/*/SKILL.md or .claude/skills/*/SKILL.md.

hashlips Bundle

Static security audit of an Agent Skill package (untrusted text only)—safe to run, data-exfil and hidden-action risks, est_tokens, and a verdict. Use only when the user explicitly asks to scan, security-scan, or sanity-check a skill.

hashlips Bundle

Authors lean, role-driven agent definitions in Markdown (responsibilities, decision framework, constraints, outputs, completion and handoff, collaboration) for use in any project or product. Use when creating `*-agent.md` files, standardizing a portable agent library, or refactoring ad-hoc prompts so they stay scannable and under a reasonable context budget.

hashlips Bundle

Compresses Markdown to minimal token form while preserving facts and agent-executable intent; duplicates sources first and compresses skill packages (SKILL.md plus reference MD). Use when reducing context cost in docs, skills, or prompts at light, medium, heavy, or extreme compression.

hashlips Bundle

Creates coherent Agent Skills with consistent format, minimal main SKILL.md files, and reference-driven documentation. Use when creating new skills, standardizing existing skills, defining skill structure, or improving skill maintainability and context efficiency.

laksh-star Bundle

Build and maintain a personal LLM Wiki: a local, interlinked markdown knowledge base compiled from raw sources. Use when the user wants to set up a knowledge wiki, ingest or compile sources into it, ask grounded questions of it, lint wiki health, find knowledge gaps, generate publishable outputs from wiki content, or export a concise context layer for a downstream assistant. Triggers include "set up my LLM wiki", "compile my wiki", "add this source to my wiki", "ask my wiki", "lint my wiki", "what gaps does my wiki have", and "export agent context".

laksh-star Bundle

Run repeatable Bumblebee supply-chain inventory scans against the current repo or explicit repo roots, including package metadata, MCP JSON configs, raw NDJSON preservation, sanitized public reports, and AI coding agent handoff notes. Use when asked to scan projects with Bumblebee, check package exposure, inventory MCP configs, run a supply-chain scan, or create public-safe dependency inventory reports.

laksh-star Bundle

Convert local .mbox email exports into local-first knowledge packs for analysis, curation, wiki compile staging, writing projects, and optional fine-tuning dataset preparation. Use when asked to analyze an mbox, summarize old emails, curate email threads, prepare a Karpathy/Claude-style wiki raw import, or derive supervised examples from an email corpus. Do not use for live mailbox operations.

laksh-star Bundle

Design, audit, and document domain-specific expert judgment distillation workflows: label taxonomies, seed datasets, prompt baselines, contested-example review, fine-tuning or Tinker-style experiment loops, failure taxonomies, repair sets, held-out evals, two-stage rubric models, and case-study writeups. Use when asked to replicate expert judgment, model a workflow after the Thinking Machines/Bridgewater Tinker article, build a domain judgment dataset, analyze a judgment model's failures, or package an experiment as a reusable case study.

laksh-star Bundle

Run or extend an agent-first competitive intelligence workflow backed by local sample data, MCP tools, optional live Tavily/OpenRouter/CocoIndex/Postgres ingestion, and generated briefs or dashboards. Use when asked to set up, demonstrate, test, package, or operate a competitive-intelligence agent; expose competitor intelligence to Claude, Codex, or other MCP clients; run sample-first demos; or switch live ingestion to arbitrary competitor names.

moussaoui-ghiles Skill

Build and tune a reusable prompt for deciding whether companies match an ideal customer profile. Use before applying AI qualification to a large company list. Test ten companies per round, incorporate user corrections, and stop only after two consecutive rounds require no corrections.

starsphere-1024 Skill

Use when the user wants to clarify or scope an ambiguous coding requirement without generating a prompt or starting implementation. Ask focused decision-blocking questions, then summarize the agreed requirement.

starsphere-1024 Bundle

Use only when the user explicitly asks for a reusable implementation prompt. Interview first, then compile the prompt. For ambiguous requirements without that request, use clarify-requirements.

dvy1987 Bundle

Design, build, validate, and ship cross-platform agent skills. Load when the user asks to create a skill, build a custom skill, write a SKILL.md, package a workflow as reusable agent capability, improve or audit a skill, publish a skill, or build a planned skill suite.